According to the CDC, healthcare-associated infections affect about two million patients in American hospitals each year. Infections like influenza and MRSA routinely spread to and among hospitalized patients, often via healthcare workers. To better understand how these infections spread, one must consider patterns of movement and interaction between elements of the population under study. Standard epidemiological models assume either that populations mix perfectly (e.g. that everyone is equally likely to encounter an infected individual), or that nonhomogenous mixing can be adequately modeled by subdividing the population into a small number of compartments according to external information (e.g., by age or type). These assumptions may not hold in small settings, where differences in a single individual's behavior can have a powerful impact on disease diffusion. Recent research in social network theory suggests a different approach to the study of infectious diseases that uses contact networks to model disease transmission. Rather than using differential equations to model diffusion of infection through a population as a group, transmission probabilities are used to determine whether or not an infected individual passes the disease on to a susceptible individual on contact. Although others have used network models to study the spread of infectious disease at an urban or regional scale, we are the only truly multidisciplinary group in the country focused on developing both network models and technology specifically to study hospital-acquired infections. The overarching goal of this research is to develop and use network epidemiology models in order to design more effective strategies for preventing and controlling the spread of healthcare-associated infections. The specific aims of this project are: (1) to create an agent-levels simulator for hospital-acquired infections based on multiple data sources and incorporating fine-grained, yet de-identified, information about person-to-person interactions;(2) to pilot and refine the technology-based acquisition, integration, and analysis of hospital movement and interaction data for the purpose of infection tracking and to support our disease simulation efforts;and (3) to apply our models to simulate diffusion of commonly-occurring hospital-acquired infections, and, more specifically, influenza and methicillin-resistant Staphylococcus aureus (MRSA). This project will provide a much-needed framework for the evaluation and comparison of hospital patient safety measures and infection control interventions such as patient cohorting, targeted hand washing compliance measures, worker vaccination strategies etc., and will prototype advanced contact-tracking technology (e.g., sensor motes) that may, in the future, be deployed proactively in the event of an infectious disease outbreak (e.g., avian influenza, SARS). In short, the methods and technology advocated here will dramatically improve our understanding of healthcare-associated infection transmission, and will translate directly to improved patient and healthcare worker safety. PUBLIC HEALTH RELEVANCE: Vaccination and hand hygiene are commonly believed to be the most effective measures for preventing the spread of hospital-acquired infections. However, we have neither a theoretical framework nor empirical data to identify workers most likely to acquire and transmit infectious agents and who therefore should have the highest priority in influenza vaccine or hand hygiene adherence campaigns. In this proposal, new methods and advanced technology are applied to model worker/patient movement and interaction to provide just such a framework;by using real healthcare worker movement data, along with real healthcare center architectural data, we are establishing a new standard for healthcare-associated disease transmission studies.